A pathological image scanning and analysis system and method

Abstract

The present invention discloses a pathological image scanning and analysis system and method, which relates to the technical field of pathological image scanning and analysis. An image scanning module is used to perform image scanning, and the scanning result is transmitted to an image segmentation module. The image segmentation module divides the overall image into multiple image sub-regions. The image data processing module first calculates and outputs the density analysis value PM of each image sub-region, and then for the image sub-regions with a density analysis value PM greater than the set density analysis value PM0 of the healthy image, calculates and outputs the abnormality degree value YQ and the final analysis result value FJ in sequence. The pathological analysis module performs direct output and iteration based on the comparative analysis of the final analysis result value FJ and the set final analysis result value FJ0 of the healthy image, and stops iterating when the final analysis result value FJ shows a continuous downward trend and enters pathological analysis. The present invention can improve the accuracy of image segmentation, enhance the sensitivity of abnormal detection, and achieve intelligent feedback adjustment.

Description

Technical Field

[0001] The present invention relates to the technical field of pathological image scanning and analysis, and in particular, to a pathological image scanning and analysis system and method. Background Art

[0002] In the field of medical image diagnosis, the analysis of pathological images is crucial for the early detection and treatment of diseases. The pathological image scanning and analysis system and method is an advanced tool developed under the background of the rapid development of medical imaging technology and artificial intelligence technology to meet the accuracy and efficiency requirements of pathological diagnosis.

[0003] Although the existing pathological image scanning and analysis systems have achieved certain results, there are still the following problems and deficiencies:

[0004] First of all, the existing image segmentation algorithms often have difficulty in accurately dividing images into regions with clear pathological significance, which affects the accuracy of subsequent analysis. Moreover, the existing anomaly detection methods often rely on simple threshold settings and are difficult to effectively detect minute anomalies in complex pathological images. In addition, there is often a lack of intelligent extended analysis for the detected pathological regions, specifically, it is unable to adaptively adjust according to the actual situation of the image, resulting in the instability of the analysis results. Summary of the Invention

[0005] The technical problem to be solved by the present invention is that there are disadvantages in the prior art such as inaccurate image segmentation, insensitive anomaly detection, and shallow and unintelligent pathological results. For this reason, we propose a pathological image scanning and analysis system and method.

[0006] The technical solution mainly is: A pathological image scanning and analysis system, including an image scanning module, an image segmentation module, an image data processing module, and a pathological analysis module. The image data processing module includes an image density analysis unit, an abnormal area analysis unit, and an analysis output and feedback unit;

[0007] The image scanning module: Scans high-resolution images by using the cooperation of image enhancement technology and image filtering technology;

[0008] The image segmentation module: Is responsible for receiving the scanned images and dividing the overall image into multiple image sub-regions;

[0009] The image data processing module: Is responsible for judging brightness and density anomalies, and calculating and outputting a density analysis value PM, an anomaly degree value YQ, and a final analysis result value FJ;

[0010] The pathological analysis module: Is responsible for iterative judgment and direct output of pathological analysis.

[0011] Preferably, the equipment used by the image scanning module includes a high-resolution medical image scanner;

[0012] The equipment used by the image segmentation module includes image processing software;

[0013] The equipment used by the image data processing module includes a computer;

[0014] The equipment used by the pathological analysis module includes pathological analysis equipment.

[0015] Preferably, the calculation formula of the image density analysis unit is as follows:

[0016] ;

[0017] ;

[0018] Where:

[0019] PM is the density analysis value, and PM reflects the average density of each image sub-region into which the overall image is divided;

[0020] LD is the brightness value, and LD reflects the average brightness of the pixels in each image sub-region;

[0021] R i is the red value of the i-th pixel, G i is the green value of the i-th pixel, B i is the blue value of the i-th pixel, and the red value R i of the i-th pixel, the green value G i of the i-th pixel, and the blue value B i of the i-th pixel respectively reflect the red, green, and blue primary color values of any pixel in each image sub-region;

[0022] YS is the number of abnormal pixels, and YS reflects the number of abnormal pixel points in each image sub-region;

[0023] ZS is the total number of pixels, and ZS reflects the total number of pixel points in each image sub-region;

[0024] In this unit, the brightness value LD calculates the brightness of the pixels in the image sub-region and is the basis for analyzing the overall density of the image. And subtracting the ratio of from 1 gives an adjustment factor, which reflects the relative quantity of normal pixel points and abnormal pixel points in the image. When there are many abnormal pixel points, the adjustment factor is small; when there are few abnormal pixel points, the adjustment factor is large.

[0025] Preferably, based on the image density analysis unit, when the density analysis value PM is greater than the set healthy image density analysis value PM0, the abnormal area analysis unit will perform calculation, and the specific calculation formula is as follows:

[0026] ;

[0027] in:

[0028] YQ is the abnormality degree value;

[0029] PM is the density analysis value;

[0030] YR is the abnormal red value;

[0031] YG is the abnormal green value;

[0032] YR max is the maximum abnormal red value, YR max Reflects the maximum degree of red abnormality present in historical pathological imaging scans;

[0033] YG max is the maximum abnormal green value, YG max Reflects the maximum degree of green abnormality present in the historical pathological imaging scan;

[0034] Maximum abnormal red value YR max and the maximum abnormal green value YG max Set to the two maximum values that have occurred in the pathological image scan, and the maximum abnormal red value YR within the set period max and the maximum abnormal green value YG max is a fixed value;

[0035] YS / 10 is the attenuation of the abnormal pixel quantity YS, thereby avoiding the instability of the abnormal degree value YQ result caused by a large number of abnormal pixels.

[0036] Preferably, the calculation formula of the analysis output and feedback unit is as follows:

[0037] ;

[0038] in:

[0039] FJ is the final analysis result value;

[0040] QZS is the total regional amount, which reflects the total amount of the image sub-regions into which the overall image is divided;

[0041] LD i is the brightness value of the i-th region;

[0042] R avg is the red mean, Ravg Reflects the average degree of red in each image sub-region into which the overall image is divided;

[0043] G avg Is the green average value, G avg Reflects the average degree of green in each image sub-region into which the overall image is divided;

[0044] R avg,max Is the maximum average value of red, R avg,max Reflects the maximum degree of red in each image sub-region into which the overall image is divided;

[0045] G avg,max Is the maximum average value of green, G avg,max Reflects the maximum degree of green in each image sub-region into which the overall image is divided;

[0046] In this unit, the ratio of ZS / YS is added to the square root of the average value of the brightness within the overall image area, and then a feedback adjustment term of the ratio of YQ / PM multiplied by the ratio of the absolute value of the difference between the average values of the red and green primary colors within the overall image area to the maximum value of the average values of the red and green primary colors is subtracted to obtain the final analysis result value FJ, where calculations are introduced for both abnormal and overall areas, presenting the degree of influence.

[0047] Preferably, the analysis based on the final analysis result value FJ and the set final analysis result value FJ0 of the healthy image is as follows:

[0048] If the final analysis result value FJ is lower than or equal to the final analysis result value FJ0 of the healthy image, it indicates that the result of the image analysis shows a low pathological degree, and there is no need to perform cyclic pathological analysis calculations;

[0049] If the final analysis result value FJ is higher than the final analysis result value FJ0 of the healthy image, it indicates that the result of the image analysis shows a high pathological degree, and cyclic pathological analysis calculations are required, and the abnormal area and the abnormal pixel amount YS are increased in the original overall image until the final analysis result value FJ shows a continuous downward trend, and the abnormal area of the final analysis result value FJ obtained in the previous calculation in the continuous downward trend is set as the pathological area that finally needs to be analyzed.

[0050] Preferably, the image density analysis unit first calculates the brightness values LD of each image sub-region, and based on the set brightness value LD0 of the healthy image, the number of brightness values LD in each corresponding region that are greater than the brightness value LD0 of the healthy image is summarized as the abnormal pixel amount YS.

[0051] There is also provided a technical solution based on the technical solution of the pathological image scanning and analysis system and having the same technical principle: A pathological image scanning and analysis method, including the following steps:

[0052] Step 1: Perform image scanning through the image scanning module and transmit the scanning result to the image segmentation module;

[0053] Step 2: Divide the overall image into multiple image sub-regions through the image segmentation module;

[0054] Step 3: First calculate and output the density analysis value PM of each image sub-region through the image data processing module, and then for the image sub-regions with a density analysis value PM greater than the set healthy image density analysis value PM0, calculate and output the abnormality degree value YQ and the final analysis result value FJ in sequence;

[0055] Step 4: Through the pathological analysis module, based on the comparative analysis of the final analysis result value FJ and the set healthy image final analysis result value FJ0, perform direct output and iteration, and stop when the final analysis result value FJ shows a continuous downward trend, and enter pathological analysis.

[0056] Technical effects and advantages of the present invention:

[0057] In the present invention, through the image segmentation introduced by the image segmentation module, accurate segmentation of pathological images can be achieved, providing an accurate basis for subsequent analysis. The method of separating for analysis and the detection of abnormal regions by the image density analysis unit after segmentation both contribute to ensuring the accuracy of pathological analysis results.

[0058] In the present invention, in the abnormal region analysis unit, by introducing the information of red-green original color difference and the number of abnormal pixel points, the detection sensitivity and specificity of abnormal regions can be significantly improved, thereby achieving accurate detection of minute abnormalities.

[0059] In the present invention, during cyclic feedback adjustment, by establishing an effective adjustment mechanism, adaptive adjustment can be performed according to the actual situation of the image, thereby optimizing the analysis result, improving the accuracy and stability of diagnosis. In addition, through cyclic feedback adjustment, the analysis result can be continuously iteratively optimized until the final analysis result value FJ continuously decreases, thereby ensuring the accuracy and reliability of the final analysis result. Description of the Drawings

[0060] Figure 1 is a flowchart of this pathological image scanning and analysis method;

[0061] Figure 2 is a structural schematic diagram of this pathological image scanning and analysis system;

[0062] Figure 3 is a structural schematic diagram of the image data processing module in the present invention. Detailed Embodiments

[0063] The present invention will now be further described in detail with reference to the accompanying drawings and preferred embodiments.

[0064] Referring Figures 1-3 As shown, the present invention provides a technical solution: a pathological image scanning and analysis system, including an image scanning module, an image segmentation module, an image data processing module, a pathological analysis module. The image data processing module includes an image density analysis unit, an abnormal area analysis unit, and an analysis output and feedback unit;

[0065] The specific analysis method process is as follows:

[0066] The image scanning module performs image scanning and transmits the scanning result to the image segmentation module;

[0067] The image segmentation module divides the overall image into multiple image sub-regions;

[0068] The image data processing module first calculates and outputs the density analysis value PM of each image sub-region, and then calculates and outputs the abnormality degree value YQ and the final analysis result value FJ in sequence for the image sub-regions with a density analysis value PM greater than the set healthy image density analysis value PM0;

[0069] The pathological analysis module performs direct output and iteration based on the comparative analysis of the final analysis result value FJ and the set healthy image final analysis result value FJ0, and stops when the final analysis result value FJ shows a continuous downward trend, and enters pathological analysis;

[0070] The equipment used by the image scanning module includes a high-resolution medical image scanner;

[0071] The equipment used by the image segmentation module includes image processing software;

[0072] The equipment used by the image data processing module includes a computer;

[0073] The equipment used by the pathological analysis module includes pathological analysis equipment.

[0074] In this embodiment, by introducing advanced algorithms and mechanisms, the problems and deficiencies existing in the prior art are solved, the accuracy and efficiency of pathological image analysis are improved, and strong support is provided for the early detection and treatment of diseases. Specifically, the image density analysis unit, the abnormal area analysis unit, and the analysis output and feedback unit each have clear calculation purposes and important significances. They jointly constitute a complete pathological image scanning and analysis system and method. Through mutual substitution and association, these formulas can achieve accurate and efficient analysis of pathological images, providing strong support for the diagnosis and treatment of diseases.

[0075] Referring Figure 1 andFigure 3 As shown in the figure, in this implementation: the calculation formula of the image density analysis unit is as follows:

[0076] ;

[0077] ;

[0078] Where:

[0079] PM is the density analysis value, and PM reflects the average density of each image sub-region into which the overall image is divided;

[0080] LD is the luminance value, and LD reflects the average luminance of the pixels in each image sub-region;

[0081] R i is the red value of the i-th pixel, G i is the green value of the i-th pixel, B i is the blue value of the i-th pixel, the red value R of the i-th pixel i the green value G of the i-th pixel i and the blue value B of the i-th pixel i respectively reflect the red, green, and blue primary color values of any pixel in each image sub-region;

[0082] YS is the abnormal pixel quantity, and YS reflects the number of abnormal pixel points in each image sub-region;

[0083] ZS is the total pixel quantity, and ZS reflects the total number of pixel points in each image sub-region;

[0084] In this unit, the luminance value LD calculates the luminance of the pixels in the image sub-region and is the basis for analyzing the overall density of the image. And subtracting the proportion of obtains an adjustment factor, and the adjustment factor reflects the relative quantity of normal pixel points and abnormal pixel points in the image. When there are many abnormal pixel points, the adjustment factor is small; when there are few abnormal pixel points, the adjustment factor is large;

[0085] The image density analysis unit first calculates the luminance value LD of each image sub-region, and based on the set healthy image luminance value LD0, sums up the number of luminance values LD greater than the healthy image luminance value LD0 in each corresponding region as the abnormal pixel quantity YS.

[0086] In the algorithm of this embodiment, Regarding the RGB values of the pixels in each sub-region of the image as a point in three-dimensional space, the distance from this point to the origin is the brightness value of the point. This method comprehensively considers the contributions of the three colors, red, green, and blue, and can more comprehensively reflect the brightness information of the pixel points. As the basis for calculating the density analysis value PM, the calculated brightness value LD reflects the light and dark degrees of the pixel points in each sub-region of the image. Moreover, by combining with the proportion of abnormal pixel points, the overall density of the image can be further analyzed;

[0087] Calculate the proportion of abnormal pixel points in the total pixel points and obtain an adjustment factor. Specifically, by taking the ratio of the amount of abnormal pixels YS to the total amount of pixels ZS, the proportion of abnormal pixel points can be obtained. Then, subtract from 1 to get an adjustment factor. This adjustment factor reflects the relative quantity of normal pixel points and abnormal pixel points in the image and is used to adjust the weight of the brightness value. The adjustment factor is used to adjust the brightness value LD so that the result of the density analysis value PM can reflect the influence of abnormal pixel points on the overall density in the image. When there are many abnormal pixel points, the adjustment factor is small, and the result of the density analysis value PM will also decrease accordingly;

[0088] The image density analysis unit of this algorithm can more accurately reflect the brightness characteristics of pixel points by calculating the brightness value LD. At the same time, by introducing the proportion of abnormal pixel points as an adjustment factor, the density analysis result is closer to the actual image situation;

[0089] Before calculating the density analysis value PM, first divide the entire image area of the case into regions. In this way, the density analysis value PM of each region can be calculated independently, which is convenient for subsequent comparison with the density analysis value PM0 of the healthy image corresponding to the same image region of a person with normal physical health. This comparison helps to quickly identify the abnormal regions in the image;

[0090] As one of the basic inputs of the abnormal region analysis unit and the analysis output and feedback unit, the accuracy of the density analysis value PM directly affects the results of subsequent abnormal region detection and loop feedback adjustment. Therefore, the beneficial effect of improving the density analysis accuracy of the image density analysis unit also provides a solid foundation for the entire pathological image scanning and analysis system.

[0091] Refer to Figure 1 and Figure 3 As shown, in this implementation: Based on the image density analysis unit, when the density analysis value PM is greater than the set density analysis value PM0 of the healthy image, it will be calculated by the abnormal region analysis unit. The specific calculation formula is as follows:

[0092] ;

[0093] Among them:

[0094] YQ is the abnormality degree value;

[0095] YR is the abnormal red value;

[0096] YG is the abnormal green value;

[0097] YR max is the maximum abnormal red value, YR max reflecting the maximum degree of red abnormality existing in historical pathological image scans;

[0098] YG max is the maximum abnormal green value, YG max reflecting the maximum degree of green abnormality existing in historical pathological image scans;

[0099] The maximum abnormal red value YR max and the maximum abnormal green value YG max are set as the two maximum values that have occurred in pathological image scans, and within the set period, the maximum abnormal red value YR max and the maximum abnormal green value YG max are fixed values;

[0100] YS / 10 is the attenuation of its abnormal pixel quantity YS, thereby avoiding the instability of the abnormality degree value YQ result caused by a large number of abnormal pixel points.

[0101] In the algorithm of this embodiment, it reflects the change of the color of abnormal pixel points in the image. In medical images, certain pathological changes will cause changes in the color of local tissues, especially the difference in the primary red and green colors. As an adjustment factor in the calculation of the abnormality degree value YQ, the absolute value of the difference in the primary red and green colors is used to enhance the detection effect of the abnormal area. When the color difference is large, the value of the adjustment factor will also increase accordingly, making the abnormality degree value YQ result more significant;

[0102] The calculation part divides the abnormal pixel quantity YS by 10, which can attenuate its quantity to a certain extent. This can avoid the abnormality degree value YQ result being too sensitive and unstable due to an excessive number of abnormal pixel points. The attenuated number of abnormal pixel points is used to adjust the abnormality degree value YQ result, making it more stable and reliable. At the same time, this attenuation also helps to balance the difference between the abnormal area and the normal area;

[0103] It reflects the relative relationship between the brighter color and the color difference in the primary red and green colors. According to the setting that max the maximum abnormal red value YR max and the maximum abnormal green value YG The result will be a constant. At this time, as the current abnormal color difference increases, that is increases, the ratio will decrease. This is because the denominator increases while the numerator remains unchanged. As an adjustment factor in the calculation of the abnormal degree value YQ, it is used to further enhance the detection effect of the abnormal area. By adjusting weights, the result of the abnormal degree value YQ can more accurately reflect the abnormal area in the image;

[0104] This algorithm unit introduces the information of the original red-green color difference, enabling the abnormal area analysis unit to more sensitively capture the color changes in the image, thereby accurately identifying the abnormal area, which is particularly important for some pathological images with inconspicuous color changes;

[0105] The abnormal area analysis unit further improves the specificity of abnormal area detection by combining the quantity information of abnormal pixel points, that is, while identifying the abnormal area, it can reduce the possibility of misjudgment, making the detection result more reliable;

[0106] The abnormal degree value YQ is one of the basic inputs of the analysis output and feedback unit, and its accuracy directly affects the results of subsequent loop feedback adjustments. Therefore, the beneficial effects of the abnormal area analysis unit in improving the detection sensitivity and specificity of the abnormal area also provide key information support for the entire pathological image scanning and analysis system;

[0107] It should be noted that the color changes in the image often can reflect the pathological changes of the tissue. Especially in medical images, the red-green difference is often related to the pathological processes of blood perfusion, inflammation, and edema. By calculating the absolute value of the original red-green color difference and the ratio with the larger value among the red and green primary colors , the abnormal area analysis unit can more sensitively capture these color changes, thereby more accurately identifying the abnormal area in the image. Introducing the calculation of the red-green difference helps to distinguish the color changes caused by pathological changes from normal physiological changes or imaging artifacts, which improves the detection specificity, that is, reduces the possibility of misjudgment, making the detection result more reliable. In addition, as an important indicator for abnormal area detection, the calculation result of the red-green difference can provide key information for subsequent analysis.

[0108] Refer to Figures 1-3 As shown, in this implementation: the calculation formula of the analysis output and feedback unit is as follows:

[0109] ;

[0110] Among them:

[0111] FJ is the final analysis result value;

[0112] QZS is the total amount of the region, and QZS reflects the total amount of the image sub-regions into which the overall image is divided;

[0113] LD i is the brightness value of the i-th region;

[0114] R avg is the average value of red, and R avg reflects the average degree of red in each image sub-region into which the overall image is divided;

[0115] G avg is the average value of green, and G avg reflects the average degree of green in each image sub-region into which the overall image is divided;

[0116] R avg,max is the maximum average value of red, and R avg,max reflects the maximum degree of red in each image sub-region into which the overall image is divided;

[0117] G avg,max is the maximum average value of green, and G avg,max reflects the maximum degree of green in each image sub-region into which the overall image is divided;

[0118] In this unit, the ratio of ZS / YS is added to the square root of the average value of the brightness within the overall image region, and multiplied by the abnormal degree value YQ and combined with After calculating the influence of the calculation part on the abnormal degree value YQ, subtract a feedback adjustment term which is the ratio of YQ / PM multiplied by the absolute value of the difference between the average values of the red and green primary colors within the overall image region and the ratio of the maximum value of the average values of the red and green primary colors, to obtain the final analysis result value FJ, where calculations are introduced for both the abnormal and overall regions, presenting the degree of influence.

[0119] In the algorithm of this embodiment, The calculation part first obtains the relative number of abnormal pixel points through the ratio of the number of abnormal pixel points to the total number of pixel points. Then, by calculating the average value of the brightness of each image sub-region and taking the square root, a value reflecting the overall brightness of the image can be obtained. Adding these two values and multiplying by the abnormal degree value YQ, a comprehensive adjustment factor is obtained, so that the result of the final analysis result value FJ can reflect the relative relationship between the abnormal region and the overall brightness in the image. By introducing the adjustment factor, the result of the final analysis result value FJ can be made more comprehensive and accurate;

[0120] Calculation part First, by calculating the ratio of the abnormality degree value YQ to the density analysis value PM, the relative relationship between the abnormal area and the overall density can be obtained. Then, by calculating the ratio of the absolute value of the difference between the average values of the red and green primary colors to the maximum value, a value reflecting the relative magnitude of the color difference can be obtained. Multiply these two values to get a feedback adjustment term, which is used to adjust the final analysis result value FJ so that it can reflect the impact of the color difference in the abnormal area on the overall analysis result in the image. By introducing the feedback adjustment term, the final analysis result value FJ can be further optimized, improving the accuracy and reliability of the analysis;

[0121] Through cyclic feedback adjustment, this algorithm unit enables the final analysis result value FJ to continuously approach the real situation, thus optimizing the analysis result. This optimization not only improves the accuracy of diagnosis but also reduces the need for human intervention and improves the analysis efficiency;

[0122] The cyclic feedback mechanism in the analysis output and feedback unit enables the system to make adaptive adjustments according to the actual situation of the image. This adaptive ability allows the system to handle different types of pathological images, improving the versatility and practicality of the system. Moreover, as the result of cyclic feedback adjustment, the accuracy of the final analysis result value FJ directly affects the depth and breadth of subsequent pathological analysis. Therefore, the beneficial effects of the analysis output and feedback unit in enhancing the system's adaptive ability and optimizing the analysis result also provide strong support for pathological analysis;

[0123] It should be noted that in the analysis output and feedback unit, the red-green color difference calculation is used to calculate the feedback adjustment term to achieve the final adjustment of image analysis. By introducing the red-green color difference information, the cyclic feedback adjustment can more accurately reflect the pathological changes in the image, thus optimizing the analysis result. The cyclic feedback mechanism in the analysis output and feedback unit enables the system to make adaptive adjustments according to the actual situation of the image. And the red-green color difference calculation, as a part of this mechanism, helps the system better handle different types of pathological images, improving the versatility and practicality of the system. By introducing the red-green color difference calculation, the analysis output and feedback unit can consider the information in the image more comprehensively, thereby enhancing the stability and accuracy of the analysis, which helps to reduce the analysis errors caused by imaging conditions and patient individual differences.

[0124] Refer to Figure 1 As shown, in this implementation: The analysis based on the final analysis result value FJ and the set final analysis result value FJ0 of the healthy image is as follows:

[0125] If the final analysis result value FJ is lower than or equal to the final analysis result value FJ0 of the healthy image, it indicates that the result of the image analysis shows a low pathological degree, and there is no need to perform cyclic pathological analysis calculations;

[0126] If the final analysis result value FJ is higher than the final analysis result value FJ0 of the healthy image, it indicates that the result of the image analysis shows a high pathological degree. When cyclic pathological analysis calculation is required, the abnormal area and the abnormal pixel amount YS are added to the original overall image until the final analysis result value FJ shows a continuous downward trend. The abnormal area of the final analysis result value FJ obtained in the previous calculation in the continuous downward trend is set as the pathological area that finally needs to be analyzed.

[0127] In this embodiment, through cyclic feedback adjustment, the final analysis result value FJ can continuously approach the real situation, which in turn affects the calculation of the density analysis value PM in the image density analysis unit. This effect helps to improve the accuracy of the density analysis value PM, making the density analysis result closer to the actual image situation. Among them, when the final analysis result value FJ is lower than and equal to the final analysis result value FJ0 of the healthy image, that is, Figure 1 when the situation of "image sub-region less than / equal to the set density analysis value of the healthy image" occurs, there is no need to enter the cyclic pathological analysis calculation again, that is, there is no need to perform the cyclic calculation from the image density analysis unit to the analysis output and feedback unit again. At this time, the final analysis result value FJ is close to the actual image situation and is within the range of the healthy image. Similarly, when the final analysis result value FJ is higher than the final analysis result value FJ0 of the healthy image, that is, Figure 1 when the situation of "image sub-region greater than the set density analysis value of the healthy image" occurs, it is necessary to add the abnormal area and the abnormal pixel amount YS to the original overall image and perform the cyclic calculation from the image density analysis unit to the analysis output and feedback unit again until the final analysis result value FJ shows a continuous downward trend;

[0128] With the continuous optimization of the final analysis result value FJ, the division and comparison of the image area by the image density analysis unit will be more accurate, which helps to quickly identify the abnormal area in the image and provides strong support for subsequent analysis;

[0129] The cyclic influence of the analysis output and feedback unit on the image density analysis unit not only improves the performance of a single formula, but also enhances the overall performance of the entire pathological image scanning and analysis system. This enhancement enables the system to more accurately identify and analyze pathological images, providing strong support for the diagnosis and treatment of diseases;

[0130] In summary, the image density analysis unit, the abnormal area analysis unit, and the analysis output and feedback unit each have significant beneficial effects, and the cyclic influence of the analysis output and feedback unit on the image density analysis unit also brings positive effects. These beneficial effects together constitute the core advantages of the pathological image scanning and analysis system, providing strong support for the diagnosis and treatment of diseases.

[0131] It should be noted that any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall also fall within the protection scope of the present invention.

Claims

1. A pathological image scanning and analysis system, characterized in that: It includes an image scanning module, an image segmentation module, an image data processing module, and a pathological analysis module. The image data processing module includes an image density analysis unit, an abnormal area analysis unit, and an analysis output and feedback unit; The image scanning module: Scans high-resolution images by cooperating with image enhancement technology and image filtering technology; The image segmentation module: Is responsible for receiving the scanned images and dividing the overall image into multiple image sub-areas; The image data processing module: Is responsible for judging brightness and density abnormalities, and calculating and outputting a density analysis value PM, an abnormality degree value YQ, and a final analysis result value FJ; The pathological analysis module: Is responsible for iterative judgment and direct output of pathological analysis; Among them, the density analysis value PM is calculated by the image density analysis unit based on the brightness value LD, the abnormal pixel amount YS, and the total pixel amount ZS. The calculation formula of the image density analysis unit is as follows: ； When the density analysis value PM is greater than the set healthy image density analysis value PM0, the abnormal area analysis unit calculates the abnormal degree value YQ based on the density analysis value PM, the abnormal red value YR, the abnormal green value YG, the maximum abnormal red value YR max , and the maximum abnormal green value YG max . The calculation formula of the abnormal area analysis unit is as follows: ； The calculation formula of the brightness value LD in the image density analysis unit is as follows: ； Where: R i is the red value of the i-th pixel, G i is the green value of the i-th pixel, B i is the blue value of the i-th pixel, and the red value R of the i-th pixel i , the green value G of the i-th pixel i , and the blue value B of the i-th pixel i respectively reflect the red, green, and blue primary color values of any pixel in each image sub-region; The density analysis value PM in the image density analysis unit reflects the average density of each image sub-area into which the overall image is divided. The brightness value LD reflects the average brightness of the pixels in each image sub-area. The abnormal pixel amount YS reflects the number of abnormal pixel points in each image sub-area. The total pixel amount ZS reflects the total number of pixel points in each image sub-area; The maximum abnormal red value YR in the abnormal area analysis unit max reflects the maximum degree of red abnormality existing in the historical pathological image scans, and the maximum abnormal green value YG max reflects the maximum degree of green abnormality existing in the historical pathological image scans; Maximum abnormal red value YR max and maximum abnormal green value YG max are set as the two maximum values that have occurred in the pathological image scan, and within the set period, the maximum abnormal red value YR max and the maximum abnormal green value YG max are fixed values; YS / 10 is the attenuation of the abnormal pixel amount YS, thereby avoiding the instability of the abnormality degree value YQ result caused by a large number of abnormal pixel points; The calculation formula of the analysis output and feedback unit is as follows: ； Where: FJ is the final analysis result value; QZS is the total amount of regions. QZS reflects the total amount of image sub-areas into which the overall image is divided; LD i is the brightness value of the i-th area; R avg is the red average value, R avg reflects the average degree of red color in each image sub-region into which the overall image is divided; G avg is the green average value, G avg reflecting the average degree of green in each image sub-region into which the overall image is divided; R avg,max is the average maximum value of red, R avg,max reflects the maximum degree of red in each image sub-region into which the overall image is divided; G avg,max is the green average maximum value, G avg,max reflecting the maximum degree of green in each image sub-region into which the overall image is divided.

2. The pathological image scanning and analysis system according to claim 1, wherein: The equipment used by the image scanning module includes a high-resolution medical image scanner; The equipment used by the image segmentation module includes image processing software; The equipment used by the image data processing module includes a computer; The equipment used by the pathological analysis module includes a pathological analysis device.

3. A pathological image scanning and analysis system according to claim 1, characterized in that: Analysis based on the final analysis result value FJ and the set final analysis result value FJ0 of the healthy image is as follows: If the final analysis result value FJ is lower than or equal to the final analysis result value FJ0 of the healthy image, it indicates that the result of the image analysis shows a low pathological degree, and there is no need to perform cyclic pathological analysis calculations; If the final analysis result value FJ is higher than the final analysis result value FJ0 of the healthy image, it indicates that the result of the image analysis shows a high pathological degree. Cyclic pathological analysis calculations need to be performed, and abnormal areas and the abnormal pixel amount YS are increased in the original overall image until the final analysis result value FJ shows a continuous downward trend. The abnormal area of the final analysis result value FJ calculated in the previous step in the continuous downward trend is set as the pathological area that finally needs to be analyzed.

4. A pathological image scanning and analysis system according to any one of claims 1-3, characterized in that: The image density analysis unit first calculates the brightness value LD of each image sub-area, and based on the set brightness value LD0 of the healthy image, sums up the number of brightness values LD in each corresponding area that are greater than the brightness value LD0 of the healthy image as the abnormal pixel amount YS.

5. An analysis method of a pathological image scanning and analysis system according to claim 1, characterized in that It includes the following steps: Step 1: Perform image scanning through the image scanning module and transmit the scanning result to the image segmentation module; Step 2: Divide the overall image into multiple image sub-regions through the image segmentation module; Step 3: Through the image data processing module, first calculate and output the density analysis value PM of each image sub-region, and then for the image sub-regions with a density analysis value PM greater than the set healthy image density analysis value PM0, calculate and output the abnormality degree value YQ and the final analysis result value FJ in sequence; Step 4: Through the pathological analysis module, based on the comparative analysis of the final analysis result value FJ and the set healthy image final analysis result value FJ0, perform direct output and iteration, and stop when the final analysis result value FJ shows a continuous downward trend, and then enter pathological analysis.

Images